| Summary: | Inconsistent use of counterName and identifier in CSS counter code and loss of information about the counter type. | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Carol Szabo <carol> | ||||
| Component: | CSS | Assignee: | Carol Szabo <carol> | ||||
| Status: | RESOLVED FIXED | ||||||
| Severity: | Normal | CC: | commit-queue, darin, webkit.review.bot | ||||
| Priority: | P2 | ||||||
| Version: | 528+ (Nightly build) | ||||||
| Hardware: | All | ||||||
| OS: | All | ||||||
| Bug Depends on: | 31723 | ||||||
| Bug Blocks: | 11031 | ||||||
| Attachments: |
|
||||||
|
Description
Carol Szabo
2009-11-23 15:57:06 PST
Created attachment 44587 [details]
Proposed Patch
Added accessors and members to the CounterNode class to differentiate between reset type nodes and nodes that act as reset because they are not in the scope of any other counter node with the same identifier. This is useful in reevaluating the values of counters when new counters are added to the DOM that may make a root node non-root.
Changed references to counterName to identifier for consistency.
style-queue ran check-webkit-style on attachment 44587 [details] without any errors.
Comment on attachment 44587 [details] Proposed Patch > + // identifier must match the identifier of this counter. > void removeChild(CounterNode*, const AtomicString& identifier); The comment here is worded strangely. You say it must match. That makes it sound like it's a programming mistake if I pass an identifier that does not match. I don't think that's how it works. Maybe this wording: "Does nothing unless the identifier matches the identifier of the counter." Or maybe you can think of something better. r=me Comment on attachment 44587 [details] Proposed Patch Clearing flags on attachment: 44587 Committed r52450: <http://trac.webkit.org/changeset/52450> All reviewed patches have been landed. Closing bug. |